CAPP(computer aided process planning) is a key technology for integration of CAD and CAM. Process planning relies on manufacturing knowledge and CAPP is characterized with multi-knowledge resources, multi-task, multi-level and multi-constrain so process planning is hard to automate. This paper introduces the artificial neural network for unstructured manufacturing knowledge modeling, knowledge is represented as neural network weight value matrix, and then form ANN database to support intelligent CAPP. Example about cutting force modification is presented to test the feasibility of this approach. As Intelligent CAPP is knowledge based and structure of CAPP varies with types of knowledge representation, this paper presents the system structure of intelligent CAPP system. This system employs the black-board inference, unified manufacturing resource and part model, multi-knowledge database to realize the process planning automatically.